nithinholla
commited on
Commit
•
c1b1430
1
Parent(s):
fe4bae9
Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,4 @@
|
|
|
|
1 |
language: nl
|
2 |
datasets:
|
3 |
- common_voice
|
@@ -49,15 +50,15 @@ resampler = torchaudio.transforms.Resample(48_000, 16_000)
|
|
49 |
# Preprocessing the datasets.
|
50 |
# We need to read the aduio files as arrays
|
51 |
def speech_file_to_array_fn(batch):
|
52 |
-
|
53 |
-
|
54 |
-
|
55 |
|
56 |
test_dataset = test_dataset.map(speech_file_to_array_fn)
|
57 |
inputs = processor(test_dataset["speech"][:2], sampling_rate=16_000, return_tensors="pt", padding=True)
|
58 |
|
59 |
with torch.no_grad():
|
60 |
-
|
61 |
|
62 |
predicted_ids = torch.argmax(logits, dim=-1)
|
63 |
|
@@ -85,30 +86,30 @@ processor = Wav2Vec2Processor.from_pretrained("nithinholla/wav2vec2-large-xlsr-5
|
|
85 |
model = Wav2Vec2ForCTC.from_pretrained("nithinholla/wav2vec2-large-xlsr-53-dutch")
|
86 |
model.to("cuda")
|
87 |
|
88 |
-
chars_to_ignore_regex = '[
|
89 |
resampler = torchaudio.transforms.Resample(48_000, 16_000)
|
90 |
|
91 |
# Preprocessing the datasets.
|
92 |
# We need to read the audio files as arrays
|
93 |
def speech_file_to_array_fn(batch):
|
94 |
-
|
95 |
-
|
96 |
-
|
97 |
-
|
98 |
|
99 |
test_dataset = test_dataset.map(speech_file_to_array_fn)
|
100 |
|
101 |
# Preprocessing the datasets.
|
102 |
# We need to read the audio files as arrays
|
103 |
def evaluate(batch):
|
104 |
-
|
105 |
|
106 |
-
|
107 |
-
|
108 |
|
109 |
-
|
110 |
-
|
111 |
-
|
112 |
|
113 |
result = test_dataset.map(evaluate, batched=True, batch_size=8)
|
114 |
|
|
|
1 |
+
---
|
2 |
language: nl
|
3 |
datasets:
|
4 |
- common_voice
|
|
|
50 |
# Preprocessing the datasets.
|
51 |
# We need to read the aduio files as arrays
|
52 |
def speech_file_to_array_fn(batch):
|
53 |
+
\tspeech_array, sampling_rate = torchaudio.load(batch["path"])
|
54 |
+
\tbatch["speech"] = resampler(speech_array).squeeze().numpy()
|
55 |
+
\treturn batch
|
56 |
|
57 |
test_dataset = test_dataset.map(speech_file_to_array_fn)
|
58 |
inputs = processor(test_dataset["speech"][:2], sampling_rate=16_000, return_tensors="pt", padding=True)
|
59 |
|
60 |
with torch.no_grad():
|
61 |
+
\tlogits = model(inputs.input_values, attention_mask=inputs.attention_mask).logits
|
62 |
|
63 |
predicted_ids = torch.argmax(logits, dim=-1)
|
64 |
|
|
|
86 |
model = Wav2Vec2ForCTC.from_pretrained("nithinholla/wav2vec2-large-xlsr-53-dutch")
|
87 |
model.to("cuda")
|
88 |
|
89 |
+
chars_to_ignore_regex = '[\\,\\?\\.\\!\\-\\;\\:\\"\\“\\%\\'\\�\\(\\)\\&\\–\\—\\—\\…\\´\\’]'
|
90 |
resampler = torchaudio.transforms.Resample(48_000, 16_000)
|
91 |
|
92 |
# Preprocessing the datasets.
|
93 |
# We need to read the audio files as arrays
|
94 |
def speech_file_to_array_fn(batch):
|
95 |
+
\tbatch["sentence"] = re.sub(chars_to_ignore_regex, '', batch["sentence"]).lower()
|
96 |
+
\tspeech_array, sampling_rate = torchaudio.load(batch["path"])
|
97 |
+
\tbatch["speech"] = resampler(speech_array).squeeze().numpy()
|
98 |
+
\treturn batch
|
99 |
|
100 |
test_dataset = test_dataset.map(speech_file_to_array_fn)
|
101 |
|
102 |
# Preprocessing the datasets.
|
103 |
# We need to read the audio files as arrays
|
104 |
def evaluate(batch):
|
105 |
+
\tinputs = processor(batch["speech"], sampling_rate=16_000, return_tensors="pt", padding=True)
|
106 |
|
107 |
+
\twith torch.no_grad():
|
108 |
+
\t\tlogits = model(inputs.input_values.to("cuda"), attention_mask=inputs.attention_mask.to("cuda")).logits
|
109 |
|
110 |
+
\tpred_ids = torch.argmax(logits, dim=-1)
|
111 |
+
\tbatch["pred_strings"] = processor.batch_decode(pred_ids)
|
112 |
+
\treturn batch
|
113 |
|
114 |
result = test_dataset.map(evaluate, batched=True, batch_size=8)
|
115 |
|